Graph World Models for Constrained Epidemic Policy Planning

2026-09-28

AI Summary

This study addresses the absence of coupled dynamic models and the difficulty of enforcing resource constraints in cross-regional epidemic planning by proposing the EpiMind framework. This approach pioneers the integration of graph-structured policy imagination with explicitly constrained coordination. Specifically, it employs a graph-decomposed recurrent state-space model to generate policy-conditioned simulations, while jointly optimizing regional interventions via a graph-temporal ADMM algorithm combined with projection techniques to strictly guarantee shared-resource feasibility. Experimental results demonstrate that EpiMind reduces the RMSE of hospitalization forecasting by 29% and achieves planning performance approaching that of the optimal feasible constant policy. Furthermore, it comprehensively outperforms existing deployable baselines in real-world scenarios.

Abstract

Epidemic policy planning often requires coordination between geographical regions, taking into account mobility-driven spillovers and how to make use of limited resources. Existing methods either lack action-conditioned models of coupled dynamics or cannot guarantee per-period feasibility. We present EpiMind, a graph world model framework for constrained epidemic policy planning across regions. A graph-factored recurrent state-space model generates joint policy-conditioned rollouts from regional latent beliefs, while graph-temporal ADMM optimizes regional interventions, enforces shared-resource feasibility through projection, and evaluates temporal specifications under the learned model. EpiMind reduces admission RMSE by 29% relative to graph-free dynamics modeling, plans within 1-5% of the best feasible constant policy with guaranteed shared-budget feasibility, and outperforms all deployable baselines across three resource budgets in real-context evaluation. These results demonstrate that graph-structured policy imagination with explicit constrained coordination supports effective epidemic interventions from learned dynamics.

Problem

• epidemic policy planning
• constrained coordination
• graph world models
• resource allocation
• coupled dynamics

Innovation

• Graph World Models
• Recurrent State-Space Model
• Graph-Temporal ADMM
• Constrained Policy Planning
• Epidemic Intervention